Microwave-Based Alkali Pretreatment of Switchgrass and Coastal Bermudagrass

for Bioethanol Production

Deepak R. Keshwani
Dept. of Biological Systems Engineering, University of Nebraska-Lincoln, Lincoln, NE 68583

Jay J. Cheng
Dept. of Biological and Agricultural Engineering, North Carolina State University, Raleigh, NC 27695

DOI 10.1002/btpr.371
Published online December 28, 2009 in Wiley InterScience (www.interscience.wiley.com).

Switchgrass and coastal bermudagrass are promising lignocellulosic feedstocks for bioe-
thanol production. However, pretreatment of lignocelluloses is required to improve produc-
tion of fermentable sugars from enzymatic hydrolysis. Microwave-based alkali pretreatment
of switchgrass and coastal bermudagrass was investigated in this study. Pretreatments were
carried out by immersing the biomass in dilute alkali reagents and exposing the slurry to
microwave radiation at 250 W for residence times ranging from 5 to 20 min. Simons’ stain
method was used to quantify changes in biomass porosity as a result of the pretreatment.
Pretreatments were evaluated based on yields of total reducing sugars, glucose, and xylose.
An evaluation of different alkalis identified sodium hydroxide as the most effective alkali re-
agent for microwave-based pretreatment of switchgrass and coastal bermudagrass. 82% glu-
cose and 63% xylose yields were achieved for switchgrass and 87% glucose and 59% xylose
yields were achieved for coastal bermudagrass following enzymatic hydrolysis of biomass
pretreated under optimal conditions. Dielectric properties for dilute sodium hydroxide solu-
tions were measured and compared with solid losses, lignin reduction, and reducing sugar
levels in hydrolyzates. Results indicate that dielectric loss tangent of alkali solutions is
a potential indicator of the severity of microwave-based pretreatments. V C 2009 American

Institute of Chemical Engineers Biotechnol. Prog., 26: 644–652, 2010

Keywords: microwave pretreatment, switchgrass, coastal bermudagrass, bioethanol, dielectric
properties

Introduction and was identified as a model energy crop by the United

Given the steady increase in energy consumption and fore-
seeable decline in crude oil production, there is considerable
interest in establishing renewable energy alternatives. In the
United States, ethanol derived from corn is being established
as an alternative to gasoline to meet liquid fuel needs. How-
ever, limitations on the availability of arable land limit the
production of corn-based ethanol and its use as a viable
replacement for gasoline consumption.1 The increased allo-
cation of corn for ethanol production raises corn prices and
could result in reduced exports of animal products due to
insufficient availability for feed applications.2 Lignocellulo-
sic materials have been identified as promising feedstock for
ethanol production. These materials include agricultural resi-
dues, newsprint, wood chips, and herbaceous energy crops
such as switchgrass and coastal bermudagrass.
Switchgrass (Panicum virgatum L.) is a perennial warm-
season grass native to North America. This feedstock shows
promise due to high productivity across a wide geographic
range, suitability for marginal land quality, low water and
nutritional requirements, and positive environmental benefits
Correspondence concerning this article should be addressed to J. J.
Cheng at jay_cheng@ncsu.edu.
States Department of Energy.3 The benefits associated with
the perennial nature of switchgrass include less intensive ag-
ricultural management practices, reduced energy and agro-
chemical consumption, and positive effects on soil and
wildlife quality.4 Biomass yield estimates for switchgrass
ranging from 7 to 15 dry tons per acre have been reported in
field trials across the southeastern United States.5–7
Coastal bermudagrass (Cynodon dactylon L.) is widely
grown in the southern United States with dry matter yields
of 6–10 tons per acre annually.8 This feedstock is grown by
many hog farmers in the southeastern United States as part
of nutrient management plans to remove nitrogen and phos-
phorous from swine wastes.9 Bermudagrass hay currently
has low-commercial value as an animal feed. Hence, utiliza-
tion of bermudagrass for ethanol would commercially benefit
hog farmers. As the agricultural practices for coastal bermu-
dagrass are already in place, it is an attractive feedstock for
bioethanol production.
The biochemical conversion of lignocellulosic materials
into ethanol involves hydrolysis of cellulose into fermentable
sugars and the subsequent fermentation of these sugars into
ethanol. The hydrolysis is carried out by carbohydrolases
that break down structural carbohydrates into monomeric
sugars such as glucose and xylose. Lignocellulosic materials

644 C 2009 American Institute of Chemical Engineers

V
Biotechnol. Prog., 2010, Vol. 26, No. 3
contain cellulose, hemicellulose, and lignin that are closely
associated in a complex crystalline structure. As a result, the
efficiency of the hydrolysis is reduced due to limited accessi-
bility of the enzymes to the substrate.10 Hence, pretreatment
processes are required to improve the efficiency of enzy-
matic hydrolysis. The purpose of pretreatment is to remove
lignin and/or hemicellulose, reduce crystallinity of cellulose,
and increase porosity of the biomass. Several physical, phys-
ico-chemical, chemical, and biological pretreatment methods
have been studied and comprehensive reviews detailing the
mode of action of these pretreatments have been
published.1,11
For switchgrass, previously examined pretreatments
include dilute sulfuric acid pretreatment,12-14 lime pretreat-
ment,15 ammonia fiber explosion (expansion),16 fractionation
using ammonia and hydrogen peroxide percolation,17,18 and
ammonia water-pretreatment.19 Recent studies on micro-
wave-based pretreatment20,21 and radio frequency-assisted al-
kali pretreatment22 have also been reported. Previous work
on bermudagrass includes high-pressure hot water pretreat-
ment,23 ammonia fiber explosion,8,24 and dilute sulfuric acid
pretreatment.9
The use of microwave radiation is a promising pretreat-
ment process that utilizes thermal and nonthermal effects
generated by microwaves in aqueous environments. The
electric and magnetic field components of microwaves apply
forces that are rapidly changing direction at the rate of 2.4
 109 times per second.25 The vibration of polar molecules
and the movement of ions results in the generation of heat
and extensive intermolecular collisions. These effects can
accelerate chemical, biological, and physical processes.26
The interaction of microwaves with any material depends on
its dielectric properties: dielectric constant and dielectric loss
factor. The dielectric constant is a measure of the ability of
a material to store electromagnetic energy and the dielectric
loss factor is a measure of the ability of a material to convert
electromagnetic energy into heat.27 Loss tangent, which is a
ratio of dielectric loss factor to the dielectric constant, is a
parameter used to describe the overall efficiency of a mate-
rial to utilize energy from microwave radiation.28 Some
advantages of microwave heating over conventional heating
include reduction of process energy requirements, uniform
and selective processing, and ability to start and stop the
process instantaneously.29 Additionally, since the heat is gen-
erated internally via direct interaction between the electro-
magnetic field and components of the heated material, the
heating is volumetric and rapid.30
The earliest known study involving microwave pretreat-
ment examined the effect of microwave radiation on rice
straw and bagasse immersed in water and reported an
improvement in total reducing sugar production by a factor
of 1.6 for rice straw and 3.2 for bagasse in comparison to
untreated biomass.31 Microwave pretreatment of sugarcane
bagasse and rice hulls soaked in water followed by lignin
extraction was reported to yield 77–84% of total available
reducing sugars.32 A similar study involving microwave pre-
treatment of rice straw and sugarcane bagasse followed by
lignin extraction reported a yield of 43–55% of total avail-
able reducing sugars.33 Microwave-based pretreatment of
rice straw soaked in dilute alkali resulted in glucose yield
and total carbohydrate conversion of 65% and 78%, respec-
tively.34 A recent study on microwave-based alkali pretreat-
ment of switchgrass reported that sugar yields with
microwave heating were higher than yields from conven-
645
tional heating.20 No work related to microwave pretreatment
of coastal bermudagrass has been reported and none of these
earlier studies have examined dielectric properties relevant
to microwave pretreatment of biomass.
The objectives of this study were to evaluate the potential
of different alkalis (calcium hydroxide, sodium carbonate,
and sodium hydroxide) for microwave-based pretreatment of
switchgrass and coastal bermudagrass, identify optimal pre-
treatment conditions for the most effective alkali reagents,
and examine the relationship between dielectric properties of
dilute alkali solutions and the effects of the pretreatment pro-
cess. It is anticipated that measuring the dielectric properties
and establishing their effect on the pretreatment process
would yield useful information for future studies to model
the process, and address scalability and process-design
issues.
Materials and Methods
Biomass
Switchgrass (oven-dried at 70 C) and coastal bermuda-
grass (sun-dried) samples were obtained from field sites at
North Carolina State University’s Central Crops Research
Station in Clayton, North Carolina. Switchgrass was har-
vested in July 2007, and coastal bermudgrass was harvested
in June 2007. The feedstock was ground to pass a 2-mm
sieve in a Thomas Wiley Laboratory Mill (Model No. 4) and
stored in sealed plastic bags at room temperature until use
for pretreatment and biomass characterization.
Pretreatment
Microwave-based pretreatments were carried out in a gen-
eral purpose laboratory microwave oven (Panasonic Corpora-
tion, model NN-S954). The apparatus provided microwave
radiation using inverter technology at variable power levels
ranging from 125 to 1,250 W. All pretreatments in this study
were carried out at a power level of 250 W. For each pre-
treatment, slurry consisting of 5 g (dry basis) of the biomass
mixed with dilute alkali reagent (solid to liquid ratio of
1:10) was exposed to microwave radiation for the desired
pretreatment time. The reaction vessel was a 500-mL glass
beaker which was placed at the center of the rotating plat-
form in the microwave. Following pretreatment, the resulting
biomass was washed with 200 mL of deionized water to
remove excess chemicals and/or other pretreatment by-prod-
ucts that might inhibit enzymatic hydrolysis. For pretreat-
ment with calcium hydroxide, low solubility prompted the
use of excess (1,000 mL) deionized water to remove resid-
ual alkali. The contents were filtered under vacuum pressure
through a medium porosity (2–5lm particle retention) quan-
titative filter paper. Moisture contents of the pretreated bio-
mass were measured after washing and samples were stored
in a sealed bag at 4 C for enzymatic hydrolysis.
Enzymatic hydrolysis
Enzymatic hydrolysis of pretreated samples was carried
out in 250-mL Erlenmeyer flasks in a controlled environment
incubator shaker (New Brunswick, model G-25) set at 55 C
and 150 rpm. One gram (dry basis) of wet pretreated bio-
mass was immersed in 50 mM sodium citrate buffer to main-
tain a pH of 4.8. Cellulase (Celluclast 1.5L) from
Trichoderma reesei (E.C. 3.2.1.4) was added at an enzyme
646 Biotechnol. Prog., 2010, Vol. 26, No. 3

loading of 40 FPU/g biomass. (Note: FPU is filter paper Table 1. Composition (wt % on a Dry Basis) of Untreated
units). The cellulase activity was supplemented with cello- Switchgrass and Coastal Bermudagrass Used in This Study
(Reported Values Are Means of Three Replicates with Standard
biase (Novozyme 188) from Aspergillus niger (E.C. Deviations in Parenthesis)
3.2.1.21) at an enzyme loading of 70 CBU/g biomass. (Note:
Component Switchgrass Bermudagrass
CBU is cellobiase units). The additional cellobiase was nec-
essary to mitigate cellobiose inhibition of cellulase.35 The Glucan 31.3 (1.4) 25.0 (1.2)
enzymes were obtained from Novozymes North America, Xylan 18.4 (0.83) 16.3 (0.76)
Arabinan 1.92 (0.42) 1.99 (0.31)
Inc. (Franklinton, NC). Before use, the activities of the Galactan 1.69 (0.25) 1.43 (0.13)
enzymes were measured using standard IUPAC procedures.36 Acid insoluble lignin 18.2 (1.5) 15.4 (1.1)
The measured activities for cellulase and cellobiase were Acid soluble lignin* 4.32 (0.34) 3.96 (0.19)
76.44 FPU/mL and 283.14 CBU/mL, respectively. Sodium Extractives 4.87 (0.94) 4.17 (1.2)
Ash 3.96 (0.82) 6.60 (1.4)
azide was added to the hydrolysis mixture at a concentration
of 0.3% w/v (weight per volume) to inhibit microbial * Acid soluble lignin was measured at 205 nm and calculated using an
growth. The hydrolysis was carried out for 72 h after which absorptivity value of 110 L/(g cm).
the hydrolyzates were centrifuged and the supernatant was
stored at 20 C for sugar analysis. The same procedure was sel fitted with a dielectric probe and temperature probe and
used during the optimization of enzyme loading for switch- sealed using a tri-clamp fitting. The vessel was heated to
grass and coastal bermudagrass pretreated under optimum desired temperatures using an oil bath (Model RTE111,
conditions. Neslan Instruments Inc., Newington, NH). The network ana-
lyzer was programmed to record dielectric properties for fre-
quencies ranging from 200 to 3,000 MHz when a desired
Biomass composition and sugar analysis temperature was reached. Dielectric properties were meas-
Structural carbohydrates and lignin content in the biomass ured for temperatures ranging from 20 to 120 C at intervals
were measured by a two-step acid hydrolysis as per a stand- of 20 C.
ard protocol from the National Renewable Energy Labora-
tory.37 Extractives in the biomass were determined by a
gravimetric method using a toluene–ethanol mixture.38 Total
reducing sugars in the hydrolyzates were measured using the Overview of experimental scheme
dinitrosalicylic acid (DNS) method.39 Monomeric sugars in
the enzymatic hydrolyzates and composition analysis liquors The effectiveness of sodium carbonate (Na2CO3), calcium
were measured with an HPLC using a refractive index detec- hydroxide (Ca(OH)2), and sodium hydroxide (NaOH) as al-
tor and a Bio-Rad Aminex HPX-87P column tailored for kali reagents for microwave-based pretreatment of switch-
separation of monosaccharides derived from lignocellulosic grass and coastal bermudagrass was determined. Biomass
materials. The column was operated at a flow rate of 0.6 samples were immersed in 2% w/v alkali solutions and
mL/min at a temperature of 80 C using HPLC grade water exposed to microwave radiation for 10 min. The pretreated
as the mobile phase. The injection volume was 10 lL and samples were enzymatically hydrolyzed and total reducing
the sample-run time was 35 min with a postrun time of 25 sugars were measured. The improvement in hydrolysis in
min between injections. comparison to untreated biomass was used to identify the
best alkali reagent(s) for each feedstock. Porosity of pre-
treated switchgrass samples was measured to explain the dif-
Measurement of biomass porosity ferences in sugar yields for pretreatment with different
The accessibility of the biomass to the hydrolytic enzymes alkalis.
was evaluated by measuring the porosity using Simons’ Stain After the identification of optimum alkali reagent(s) for
method40,41 which is based on the competitive adsorption of the pretreatment of switchgrass and coastal bermudagrass, a
two dyes: Direct Orange 15 and Direct Blue 1. The large or- factorial experiment was designed in each case to study the
ange dye molecules can only penetrate large pores in the effects of concentration and pretreatment time and any
biomass. They replace the smaller blue dye molecules in potential interactions. The experimental design consisted of
these larger pores because of greater affinity for hydroxyl three levels of alkali concentration (1%, 2%, and 3% w/v)
groups of cellulose.42 The numbers of both small and large and four levels of pretreatment time (5, 10, 15, and 20 min).
pores are important factors in determining the accessibility Comparisons between pretreatments were based on yields of
of cellulase to the biomass.43 The amount of orange and total reducing sugars, glucose and xylose after accounting
blue dyes adsorbed by the biomass can be used to quantity for solid losses during pretreatment.
the relative difference in porosity between biomass sam-
ples.44 The blue dye has a molecular diameter of 1 nm
which is significantly less than the molecular diameter of the
orange dye which ranges from 5 to 7 nm.41,42,44
Statistical analysis
All experiments in this study were conducted in triplicate
Measurement of dielectric properties and statistical analysis was done using SAS 9.1.3 (SAS Insti-
Dielectric properties of the chemical reagents used to soak tute Inc., Cary, NC). Error bars shown in all figures are
the biomass were measured using an open-ended coaxial based on standard deviation values. Comparisons between
dielectric probe (Model HP 85070B, Agilent Technologies, treatment means were carried out using the Tukey adjust-
Palo Alto, CA) connected to a network analyzer (Model HP ment under PROG GLM functionality. The Tukey adjust-
8753C, Agilent Technologies) based on a previously reported ment enables comparison of multiple treatments while
procedure.27 The solutions were placed in a cylindrical ves- maintaining the probability of Type I error (a) at 0.05.
Biotechnol. Prog., 2010, Vol. 26, No. 3 647

Figure 1. Total reducing sugar production from enzymatic hydrolysis of untreated biomass and washed microwave-pretreated
biomass using different alkali reagents.

Results and Discussion

Biomass characterization
Chemical composition of the feedstocks used in this study
is shown in Table 1. Although the amounts of minor carbo-
hydrates (arabinan and galactan) were similar for both feed-
stocks, switchgrass had significantly higher glucan, xylan,
and lignin. Using anhydro corrections (0.88 for six carbon
sugars and 0.9 for five carbon sugars), the maximum
amounts of glucose and xylose from switchgrass theoreti-
cally available for conversion into ethanol are 355 mg/g bio-
mass and 204 mg/g biomass, respectively. For bermudagrass,
these amounts are 284 mg/g biomass for glucose and 181
mg/g biomass for xylose. These theoretically available
amounts of sugars were the basis for the %sugar yields
reported in this study.

Evaluation of different alkali reagents

Three different alkali reagents were evaluated (Na2CO3,
Ca(OH)2, and NaOH) for the pretreatment of switchgrass
and coastal bermudagrass using 2% w/v alkali concentration
and 10 min pretreatment time with a solid to liquid ratio of
1:10. Reducing sugar yields (based on raw biomass) from
enzymatic hydrolysis of pretreated biomass was compared
with untreated biomass (control) (Figure 1). For switchgrass,
reducing sugar yields for microwave pretreatment with
Na2CO3, Ca(OH)2, and NaOH were 207, 372, and 446 mg/g.
These yields are significantly and substantially higher than
the yield from untreated switchgrass (151 mg/g). For coastal
bermudagrass, reducing sugar yields for microwave pretreat-
ment with Na2CO3, Ca(OH)2, and NaOH were 204, 274, and
424 mg/g, respectively. These yields are significantly and
substantially higher than the yield from untreated bermuda-
grass (119 mg/g). For both feedstocks, pretreatment with
NaOH resulted in significantly higher reducing sugar yields
over pretreatment with Na2CO3 and Ca(OH)2. Therefore,
NaOH was chosen for further study to optimize the concen-
tration and time for both switchgrass and bermudagrass.
To explain the differences in the performance of Na2CO3,
Ca(OH)2, and NaOH as an alkali reagent for microwave pre-
treatment, the porosity of pretreated switchgrass was quanti-
fied using the Simons’ Stain method. The amounts of the
smaller blue dye and the larger orange dye adsorbed by the
biomass indicate the extent to which the accessibility to cel-
lulase improved because of the pretreatment. A recent study
reported a correction coefficient (r2) of 0.95 between adsorp-
tion of Simons’ stain dye and hydrolysis yields.45
Figure 2a shows the adsorption of the blue dye by pre-
treated switchgrass. The extent of absorption was signifi-
Figure 2. Amount of dye adsorbed by untreated and micro-
wave-pretreated switchgrass using different alkali
reagents (a) blue dye (molecular diameter of 1 nm)
and (b) orange dye (molecular diameter of 5–7 nm).
cantly higher for biomass treated with all alkalis in
comparison to untreated switchgrass. Dye adsorption was
statistically similar for biomass pretreated with Na2CO3 and
Ca(OH)2 (59 and 62 mg/g, respectively). The highest adsorp-
tion was observed for biomass pretreated with NaOH (82
mg/g). Figure 2b shows the adsorption of the orange dye by
pretreated switchgrass. Like the blue dye, the extent of
absorption of the orange dye was significantly higher for bio-
mass treated with alkalis than for untreated switchgrass.
However, biomass pretreated with Na2CO3 adsorbed substan-
tially lower orange dye than biomass pretreated with
Ca(OH)2 (8.3 and 16 mg/g, respectively). The adsorption of
orange dye was the highest for NaOH (32 mg/g).
From Figure 2, it is observed that the relative differences
in dye adsorption between different alkali pretreatments are
more pronounced for the orange dye. This is not surprising
since one of the purposes of pretreatment is the disruption of
the biomass structure to increase porosity. A stronger alkali
like NaOH will result in a more severe pretreatment causing
more disruption to the biomass structure and creating a
higher number of large pores. The molecular diameter of the
orange dye (5–7 nm) is comparable to reported dimensions
of the catalytic core of endoglucanase from Trichoderma
648
Figure 3. Solid loss and lignin reduction during microwave
pretreatment of switchgrass using NaOH.
reesei (4  5  6 nm).41 Since the orange dye is adsorbed
into the larger pores, it is reasonable to expect the amount
adsorbed to more accurately reflect the accessibility of the
biomass to the enzyme.
Microwave pretreatment of switchgrass using NaOH
Figure 3a shows the solid losses for pretreatment of
switchgrass using NaOH at the conditions studied in the ex-
perimental design. Solid losses ranged from 25% for pre-
treatment with 1% NaOH for 5 min to 57% for pretreatment
with 3% NaOH for 20 min. The overall effects of concentra-
tion and time were significant. An increase in the severity of
the pretreatment by increasing NaOH concentration or pre-
treatment time resulted in significantly higher solid losses.
Statistical analysis also included some noteworthy interaction
effects that elucidate the effect of concentration and time in
determining pretreatment severity. Changing pretreatment
conditions by increasing one factor while simultaneously
decreasing the other has no effect on solid losses during pre-
treatment. For example, solid loss at 2% NaOH and 5 min
was statistically similar to solid loss at 1% NaOH and 10
min. Such similarity was also observed for the following
pairs of pretreatment conditions: 1%, 15 min and 2%, 10
min; 1%, 20 min and 2%, 15 min; 2%, 10 min and 3%, 5
min; 2%, 15 min and 3%, 10 min; and 2%, 20 min and 3%
15 min.
Figure 3b shows the lignin reduction caused by the pre-
treatment of switchgrass using NaOH at the conditions stud-
ied in the experimental design. Lignin reduction ranged from
58% for pretreatment with 1% NaOH for 5 min to 83% for
pretreatment with 3% NaOH for 20 min. The overall trends
were similar to those observed for solid loss. The effects of
both alkali concentration and pretreatment time were signifi-
cant. Higher concentration and longer pretreatments resulted
in higher lignin reduction. The interaction effects were simi-
lar to those observed for solid loss. This further supports the
contention that both alkali concentration and pretreatment
time influence the severity of the pretreatment process.
Biotechnol. Prog., 2010, Vol. 26, No. 3
Figure 4. Sugar production from enzymatic hydrolysis of
microwave-pretreated switchgrass using NaOH (a)
glucose and (b) xylose.
Figure 4a shows the glucose yield from switchgrass at dif-
ferent pretreatment conditions of the experimental design.
The highest yield (289 mg/g) was obtained for pretreatment
with 2% NaOH for 10 min. For 1% and 2% NaOH, peak
yields were obtained at pretreatment time of 10 min. How-
ever, for 3% NaOH, glucose yield continually decreased
with increasing pretreatment time due to higher solid loss.
The effect of alkali concentration on glucose yields also var-
ied with pretreatment time. For a pretreatment time of 5
min, glucose yield significantly increased with increasing al-
kali concentration. At all other pretreatment times (10 min
and longer), glucose yield peaked at 2% NaOH with signifi-
cantly lower yields at 1% and 3% NaOH.
Figure 4b shows the xylose yield from switchgrass at dif-
ferent pretreatment conditions of the experimental design.
The highest yield (127 mg/g) was obtained for pretreatment
with 2% NaOH for 10 min. The overall effect of pretreat-
ment time on xylose yields was similar to that observed for
glucose yields. For 1% and 2% NaOH, peak xylose yields
were obtained at 10 min. For 3% NaOH, xylose also contin-
ually decreased with as pretreatment time was increased.
However, this decrease in xylose yield at 3% NaOH was
more pronounced than the decrease in glucose yield at 3%
NaOH. The effect of alkali concentration on xylose yields
also varied with pretreatment time. For a pretreatment time
of 5 min, alkali concentration was not a significant factor. At
longer pretreatment times (10 min and more), xylose yields
were highest at 2% NaOH. One exception to note is at 15
min, where xylose yields at 1% and 2% were not signifi-
cantly different.
Microwave pretreatment of coastal bermudagrass
using NaOH
Figure 5a shows the solid losses for pretreatment of
coastal bermudagrass using NaOH at the conditions studied
in the experimental design. Solid losses ranged from 32%
for pretreatment with 1% NaOH for 5 min to 59% for pre-
treatment with 3% NaOH for 20 min. The overall effects of
Biotechnol. Prog., 2010, Vol. 26, No. 3
Figure 5. Solid loss and lignin reduction during microwave
pretreatment of coastal bermudagrass using NaOH.
concentration and time were significant and trends were sim-
ilar to those observed for switchgrass. An increase in the se-
verity of the pretreatment by increasing NaOH concentration
or pretreatment time generally resulted in significantly higher
solid losses. An exception occurs for a pretreatment time of
5 min, where the difference between solid losses at 1% and
2% NaOH was not significant.
Figure 5b shows the lignin reduction caused by the pre-
treatment of coastal bermudagrass using NaOH at the condi-
tions studied in the experimental design. Lignin reduction
ranged from 62% for pretreatment with 1% NaOH for 5 min
to 81% for pretreatment with 3% NaOH for 20 min. The
overall trends were similar to those observed for lignin
reduction of switchgrass. Higher concentration and longer
pretreatments resulted in higher lignin reduction. An excep-
tion occurs for a pretreatment time of 5 min, where the dif-
ference between lignin reduction at 1% and 2% NaOH was
not significant. The interaction effects were similar to those
observed for lignin reduction of switchgrass.
Figure 6a shows the glucose yield from coastal bermuda-
grass at different pretreatment conditions of the experimental
design. The highest yield (252 mg/g) was obtained for pre-
treatment with 2% NaOH for 10 min but was statistically
similar to the glucose yield at 1% NaOH and 10 min (248
mg/g). An additional experiment at 0.5% NaOH and 10 min
resulted in significantly lower glucose yield (198 mg/g).
Sugar yields increased as the pretreatment time was extended
from 5 to 10 min and then decreased at longer pretreatment
times. The effect of alkali concentration on glucose yields
also varied with pretreatment time. At pretreatment times of
5 min and 10 min, glucose yield was not significantly influ-
enced by alkali concentration. The concentration effect
becomes significant at pretreatment times of 15 and 20 min.
Figure 6b shows the xylose yield from coastal bermuda-
grass at different pretreatment conditions of the experimental
design. The highest yield (107 mg/g) was obtained for pre-
treatment with 1% NaOH for 10 min. An additional experi-
ment at 0.5% NaOH and 10 min resulted in significantly
lower xylose yield (86.1 mg/g). The overall trend for the
effect of time was similar to the trend observed for xylose
649
Figure 6. Sugar production from enzymatic hydrolysis of
microwave-pretreated coastal bermudagrass using
NaOH (a) glucose and (b) xylose.
production from switchgrass. For 1% and 2% NaOH sugars,
peak xylose yields were obtained at 10 min. For 3% NaOH,
xylose also continually decreased as pretreatment time was
increased. The effect of alkali concentration on xylose yields
did not vary with pretreatment time. In general, a decrease
in alkali concentration resulted in reduced xylose yields. A
notable exception is at a pretreatment time of 10 min where
xylose yields at 1% and 2% NaOH were not significantly
different.
Differences between switchgrass and coastal bermudagrass
During the study, some differences between switchgrass
and coastal bermudagrass were observed. The most impor-
tant difference observed was the NaOH requirements for op-
timum pretreatment conditions. Switchgrass required 2%
NaOH and coastal bermudagrass required 1% NaOH. This
could be explained by the lower amount of lignin present in
bermudagrass in comparison to switchgrass (19% vs. 22%).
Another difference noted between switchgrass and coastal
bermudagrass was the overall sugar yields. For switchgrass,
glucose yields increased from 30 to 82% and xylose yields
increased from 9 to 63% after optimal pretreatment. For
coastal bermudagrass, glucose yields increased from 34 to
87% and xylose yields increased from 7 to 60% after opti-
mal pretreatment. Although the xylose yields for both feed-
stocks are similar, the glucose yields are higher for
bermudagrass (raw and pretreated). To explain this, the po-
rosity of the untreated and optimally pretreated feedstock
was quantified using the Simons’ stain method. Figures 7a,b
show the amount of blue (smaller) dye and orange (larger)
dye adsorbed by untreated switchgrass and coastal bermuda-
grass. Untreated coastal bermudagrass adsorbed 30% more
blue dye and 50% more orange dye in comparison to
untreated switchgrass. This indicates that untreated coastal
bermudagrass is more porous than untreated switchgrass and
is therefore more susceptible to enzymatic hydrolysis. A sim-
ilar result was observed when the Simons’ stain method was
used on optimally pretreated biomass (Figures 7c,d).
650
Figure 7. Amount of dye adsorbed by untreated (raw) and
microwave-pretreated (under optimal conditions)
biomass (a) blue dye adsorbed by raw switchgrass
(Raw SG) and raw coastal bermudagrass (Raw BG),
(b) orange dye adsorbed by Raw SG Raw BG, (c)
blue dye adsorbed by pretreated switchgrass (Pret.
SG) and pretreated coastal bermudagrass (Pret.
BG), and (d) orange dye adsorbed by Pret.
SG and Pret. BG.
Figure 8. Dielectric loss tangents of dilute NaOH solutions at
different temperatures.
Pretreated coastal bermudagrass adsorbed 28% more blue
dye and 50% more orange dye in comparison to pretreated
switchgrass. This indicates that pretreated coastal bermuda-
grass is more porous than pretreated switchgrass and is
therefore more susceptible to enzymatic hydrolysis.
Dielectric properties of NaOH
Dielectric properties of dilute NaOH solutions were meas-
ured for temperatures ranging from 20 to 120 C at micro-
wave frequencies from 200 to 3,000 MHz. The data set at
2,450 MHz was isolated. The dielectric loss tangents for 1%,
2%, and 3% w/v NaOH solutions are shown in Figure 8.
The loss tangent increases in value with increasing concen-
tration and the relative differences are more significant at
higher temperatures. Therefore, solutions of higher alkali
concentration are more effective in absorbing microwave
energy and converting the stored energy into heat. Since the
conduction of ions is a major factor for thermal and nonther-
mal effects in microwave processes, a system with a higher
Biotechnol. Prog., 2010, Vol. 26, No. 3
Figure 9. Relationship between dielectric loss tangent and (a)
solid loss, (b) lignin reduction, and (c) biomass
hydrolyzability at different pretreatment times.
concentration of NaOH will exhibit more ionic conduction in
response to microwave pretreatment.
During the pretreatment experiments, it was observed that
the average bulk temperature of the slurry during micro-
wave-based alkali pretreatment at 250 W was between 80 to
85 C. Dielectric loss tangents at 80 C corresponding to the
appropriate NaOH concentrations were compared with solid
losses (Figure 9a), lignin reduction (Figure 9b), and biomass
hydrolyzability (Figure 9c) from microwave pretreatment of
switchgrass using NaOH. Solid losses and lignin reduction
increase as loss tangent increases at all pretreatment times.
However, biomass hydrolyzability (expressed as mg of
reducing sugars per g of pretreated biomass) does not exhibit
this relationship with the loss tangent. This indicates that
dielectric properties of alkali solutions are a reasonable indi-
cator of pretreatment severity. However, they are not indica-
tive of biomass hydrolyzability, which is likely to be
significantly influenced by the presence of residual lignin
and crystallinity index of cellulose in the pretreated biomass.
Comparison with other studies
For switchgrass, optimal pretreatment with 2% NaOH for
10 min resulted in the highest sugar yields with 82% for glu-
cose and 63% for xylose. Yields for microwave-based pre-
treatment of switchgrass using NaOH compare well with
other studies on pretreatment of switchgrass (Table 2).
AFEX pretreatment of switchgrass reported a higher glucose
Biotechnol. Prog., 2010, Vol. 26, No. 3
Table 2. Sugar Yields from Different Studies on Pretreatment of Switchgrass and Coastal Bermudagrass
Biomass Pretreatment % Glucose Yield
Switchgrass Lime (Ca(OH)2)
Ammonia
AFEX
Dilute sulfuric acid
RF heating-NaOH
Microwave-NaOH
Microwave-NaOH
Bermudagrass AFEX
Dilute sulfuric acid
Microwave-NaOH
yield (93%) but ammonia requirements noted in the study
were high (1 kg of ammonia per kg of dry biomass).16 Glu-
cose yields are comparable to studies on NaOH pretreatment
of switchgrass using microwave radiation and radio fre-
quency heating, but xylose yields were lower since both
these studies accounted for the xylose solubilized into the
pretreatment liquor when determining overall sugar
yields.20,22
For coastal bermudagrass, optimal pretreatment with 1%
NaOH for 10 min increased glucose yields from 34 to 87%
and increased xylose yields from 7 to 59% compared with
untreated bermudagrass. Glucose yields compare favorably
with results reported from dilute acid pretreatment9 and
AFEX pretreatment8 (data shown in Table 2). However, the
dilute acid pretreatment study9 reported 81% xylose yield in
the prehydrolyzate, which is not surprising since the primary
mode of action of dilute acid pretreatment is solubilization
of hemicelluloses into the prehydrolyzate.1,11 In the case of
alkaline pretreatment, the mode of action is the removal of
lignin, which improves the enzymatic hydrolysis of the poly-
saccharides that remain.11
Conclusions
Microwave-based pretreatment of switchgrass and coastal
bermudagrass was investigated to improve sugar yields from
enzymatic hydrolysis. A comparison of different alkalis
determined that pretreatment with NaOH gave the highest
total reducing sugar yields for microwave-based pretreatment
of both feedstocks. Significant differences in the porosity of
the biomass pretreated with different alkalis were observed
using the Simons’ stain method. For switchgrass, optimal
pretreatment with 2% NaOH for 10 min resulted in yields of
82% for glucose and 63% for xylose. For coastal bermuda-
grass, optimal pretreatment with 1% NaOH for 10 min
resulted in yields of 87% for glucose and 59% for xylose.
Based on NREL’s theoretical ethanol yield calculator, yields
were calculated for both feedstocks. For pretreated switch-
grass, theoretical ethanol yields were 44 gallons per dry ton
(based on fermentation of only glucose) and 64 gallons per
dry ton (based on fermentation of both glucose and xylose).
For pretreated coastal bermudagrass, the corresponding theo-
retical ethanol yields were 38 (glucose) and 55 (glucose and
xylose) gallons per dry ton.
Dielectric properties of dilute NaOH solutions were meas-
ured and compared with solid losses, lignin reduction, and
biomass hydrolyzability. It was noted that dielectric loss tan-
gents increase with increasing NaOH concentration. Further-
more, it was determined that loss tangents are a reasonable
indicator of pretreatment severity in terms of solid loss and
lignin reduction, but do not accurately reflect the sugar pro-
duction potential of pretreated biomass.
651
% Xylose Yield Reference Number
58 58 15
45–56 25–31 19
93 [90 16
75–84 Not reported 13
85 84 22
78–88 87–96 20
82 63 This study
63 Not reported 8
52 81 9
87 59 This study
This study has shown the feasibility of microwave-based
alkali pretreatment of lignocellulosic materials to improve
yields of fermentable sugars available for ethanol production.
However, the experiments were conducted in a laboratory-
scale batch microwave system. Currently, most industrial
applications involving microwave-based processing use con-
tinuous systems. Therefore, the next logical step is to verify
the feasibility of this pretreatment method using laboratory
scale continuous microwave systems. This will enable the
identification of appropriate process parameters that will be
useful in the scale-up of microwave-based pretreatment. In
addition, future studies should also address scalability con-
siderations such as capital cost, energy consumption, and
water requirements for the process.
Acknowledgments
The authors thank the Golden LEAF Foundation of North
Carolina and the North Carolina Biotechnology Center for
funding this study, Dr. Joseph Burns from the Crop Science
Dept. of North Carolina State University for providing the bio-
mass, Dr. K.P. Sandeep from the Food Science Dept. of North
Carolina State University for access to analytical equipment,
and Novozymes North America (Franklinton, North Carolina,
USA) for providing the enzymes used in this study.
Literature Cited
1. Sun Y, Cheng J. Hydrolysis of lignocellulosic materials for
ethanol production: a review. Bioresour Technol. 2002;83:1–11.
2. Elobeid A, Tokgoz S, Hart C. The ethanol outlook for Brazil
and the United States and implications for livestock. Int Sugar
J. 2007;109:174–177.
3. McLaughlin SB. New switchgrass biofuels research program for
the southeast. Proceedings of Technical Meeting, Dearborn, MI.
Nov. 2–5, 1992. Ann Aut Techn. 1993:111–115.
4. Dunn CP, Stearns F, Guntenspergen GR, Sharpe DM. Ecologi-
cal benefits of the conservation reserve program. Conserv Biol.
1993;7:132–139.
5. Parrish DJ, Fike JH. The biology and agronomy of switchgrass
for biofuels. Crit Rev Plant Sci. 2005;24:423–459.
6. Sanderson MA, Reed RL, McLaughlin SB, Wullschleger SD,
Conger BV, Parrish DJ, Wolf DD, Taliaferro C, Hopkins AA,
Ocumpaugh WR, Hussey MA, Read JC, Tischler CR. Switch-
grass as a sustainable bioenergy crop. Bioresour Technol. 1996;
56: 83–93.
7. Thomason WE, Raun WR, Johnson GV, et al. Switchgrass
response to harvest frequency and time and rate of applied
nitrogen. J Plant Nutr. 2004;27:1199–1226.
8. Holtzapple MT, Ripley EP, Nikolaou M. Saccharification, fer-
mentation, and protein recovery from low-temperature afex-
treated coastal bermudagrass. Biotechnol Bioeng. 1994;44:1122–
1131.
9. Sun Y, Cheng JJ. Dilute acid pretreatment of rye straw and ber-
mudagrass for ethanol production. Bioresour Technol. 2005;96:
1599–1606.
652
10. McMillan JD. Pretreatment of lignocellulosic biomass. In: En-
zymatic Conversion of Biomass for Fuels Production, Himmel
ME, Baker JO, Overend RP., Editors. Vol. 566. 1994:292–
324.
11. Mosier N, Wyman C, Dale B, et al. Features of promising tech-
nologies for pretreatment of lignocellulosic biomass. Bioresour
Technol. 2005;96:673–686.
12. Chung YC, Bakalinsky A, Penner MH. Enzymatic saccharifica-
tion and fermentation of xylose-optimized dilute acid-treated
lignocellulosics. Appl Biochem Biotechnol. 2005;121:947–961.
13. Dien BS, Jung HJG, Vogel KP, et al. Chemical composition
and response to dilute-acid pretreatment and enzymatic saccha-
rification of alfalfa, reed canarygrass, and switchgrass. Biomass
Bioenergy. 2006;30:880–891.
14. Fenske JJ, Esteghlalian A, Mckinnis J, Hashimoto AG, Penner
MH. Optimization of dilute-acid pretreatment of switchgrass.
Abstr Pap Am Chem Soc. 1994;207:70-BIOT.
15. Chang VS, Burr B, Holtzapple MT. Lime pretreatment of
switchgrass. Appl Biochem Biotechnol. 1997;63:3–19.
16. Alizadeh H, Teymouri F, Gilbert TI, Dale BE. Pretreatment of
switchgrass by ammonia fiber explosion (AFEX). Appl Biochem
Biotechnol. 2005;121:1133–1141.
17. Iyer PV, Wu ZW, Kim SB, Lee YY. Ammonia recycled perco-
lation process for pretreatment of herbaceous biomass. Appl
Biochem Biotechnol. 1996;57:121–132.
18. Kim SB, Lee YY. Fractionation of herbaceous biomass by am-
monia-hydrogen peroxide percolation treatment. Appl Biochem
Biotechnol. 1996;57:147–156.
19. Kurakake M, Kisaka W, Ouchi K, Komaki T. Pretreatment with
ammonia water for enzymatic hydrolysis of corn husk, bagasse,
and switchgrass. Appl Biochem Biotechnol. 2001;90:251–259.
20. Hu ZH, Wen ZY. Enhancing enzymatic digestibility of switch-
grass by microwave-assisted alkali pretreatment. Biochem Eng
J. 2008;38:369–378.
21. Keshwani DR, Cheng JJ, Burns JC, Li L, Chiang V. Microwave
pretreatment of switchgrass to enhance enzymatic hydrolysis.
2007 ASABE Annual Meeting, Minneapolis, MN. Jun. 17–20,
2007. Paper number 077127.
22. Hu ZH, Wang YF, Wen ZY. Alkali (NaOH) pretreatment of
switchgrass by radio frequency-based dielectric heating. Appl
Biochem Biotechnol. 2008;148:71–81.
23. Brandon SK, Eiteman MA, Patel K, et al. Hydrolysis of Tifton
85 bermudagrass in a pressurized batch hot water reactor. J
Chem Technol Biotechnol. 2008;83:505–512.
24. Delarosa LB, Reshamwala S, Latimer VM, Shawky BT, Dale
BE, Stuart ED. Integrated production of ethanol fuel and protein
from coastal bermudagrass. Appl Biochem Biotechnol. 1994;45:
483–497.
25. Galema SA. Microwave chemistry. Chem Soc Rev. 1997;
26:233–238.
26. Sridar V. Microwave radiation as a catalyst for chemical reac-
tions. Curr Sci. 1998;74:446–450.
27. Kumar P, Coronel P, Simunovic J, Truong VD, Sandeep KP.
Measurement of dielectric properties of pumpable food materi-
als under static and continuous flow conditions. J Food Sci.
2007;72:E177–E183.
28. Nelson SO, Datta CJ. Dielectric properties of food materials
and electric field interactions. In: Datta AK, Anantheswaran
Biotechnol. Prog., 2010, Vol. 26, No. 3
RC, editors. Handbook of Microwave Technology for Food
Applications. New York: Marcel Dekker Inc.; 2001:69–107.
29. Datta AK. Fundamentals of heat and moisture transport for
microwaveable food product and process development. In: Datta
AK, Anantheswaran RC, editors. Handbook of Microwave Tech-
nology for Food Applications. New York: Marcel Dekker Inc.;
2001:115–166.
30. de la Hoz A, Diaz-Ortiz A, Moreno A. Microwaves in organic
synthesis. Thermal and non-thermal effects. Chem Soc Rev.
2005;34:164–178.
31. Ooshima H, Aso K, Harano Y, Yamamoto T. Microwave treat-
ment of cellulosic materials for their enzymatic-hydrolysis. Bio-
technol Lett. 1984;6:289–294.
32. Azuma JI, Tanaka F, Koshijima T. Enhancement of enzymatic
susceptibility of lignocellulosic wastes by microwave irradia-
tion. J Ferment Technol. 1984;62:377–384.
33. Kitchaiya P, Intanakul P, Krairiksh M. Enhancement of enzy-
matic hydrolysis of lignocellulosic wastes by microwave pre-
treatment under atmospheric-pressure. J Wood Chem Technol.
2003;23:217–225.
34. Zhu SD, Wu YX, Yu ZN, Liao JT, Zhang Y. Pretreatment by
microwave. Process Biochem. 2005;40:3082–3086.
35. Ryu DDY, Mandels M. Cellulases—biosynthesis and applica-
tions. Enzyme Microb Technol. 1980;2:91–102.
36. Ghose TK. Measurement of cellulase activities. Pure Appl
Chem. 1987;59:257–268.
37. Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton
D. Determination of structural carbohydrates and lignin in bio-
mass. Laboratory analytical procedure, NREL/TP-510-42618.
2007; Available at: http://www.nrel.gov/biomass/analytical_ pro-
cedures.html. First accessed Oct. 10, 2007.
38. Han J, Rowell J. Chemical composition of agro-based fibers.
In: Rowell R, Young AR, Rowell J, editors. Paper and Compo-
sites from Agro-Based Resources. New York: CRC Lewis Pub-
lishers; 1997:81–134.
39. Miller GL. Use of dinitrosalicylic acid reagent for determination
of reducing sugar. Anal Chem. 1959;31:426–428.
40. Simons FL. A stain for use in the microscopy of beaten fibers.
Tappi J. 1950;33:312–314.
41. Esteghlalian AR, Bilodeau M, Mansfield SD, Saddler JN. Do
enzymatic hydrolyzability and Simons’ stain reflect the changes
in the accessibility of lignocellulosic substrates to cellulase
enzymes? Biotechnol Prog. 2001;17:1049–1054.
42. Yu XC, Minor JL, Atalla RH. Mechanism of action of Simons’
stain. Tappi J. 1995;78:175–180.
43. Tanaka M, Ikesaka M, Matsuno R, Converse AO. Effect of
pore-size in substrate and diffusion of enzyme on hydrolysis of
cellulosic materials with cellulases. Biotechnol Bioeng. 1988;
32:698–706.
44. Yu XC, Atalla RH. A staining technique for evaluating the pore
structure variations of microcrystalline cellulose powders. Pow-
der Technol. 1998;98:135–138.
45. Chandra R, Ewanick S, Hsieh C, Saddler JN. The characteriza-
tion of pretreated lignocellulosic substrates prior to enzymatic
hydrolysis, part 1: a modified Simons’ staining technique. Bio-
technol Prog. 2008;24:1178–1185.
Manuscript received May 21, 2009, and revision received Oct. 6, 2009.